WeightTransmitter: Weighted Association Rule Mining Using Landmark Weights
Weighted Association Rule Mining (WARM) is a technique that is commonly used to overcome the well-known limitations of the classical Association Rule Mining approach. The assignment of high weights to important items enables rules that express relationships between high weight items to be ranked ahead of rules that only feature less important items. Most previous research to weight assignment has used subjective measures to assign weights and are reliant on domain specific information. Whilst there have been a few approaches that automatically deduce weights from patterns of interaction between items, none of them take advantage of the situation where weights of only a subset of items are known in advance. We propose a model, WeightTransmitter, that interpolates the unknown weights from a known subset of weights.
KeywordsWeight Estimation Landmark Weights Association Rule Mining
Unable to display preview. Download preview PDF.
- 1.Brijs, T., Swinnen, G., Vanhoof, K., Wets, G.: Using association rules for product assortment decisions: A case study. In: Knowledge Discovery and Data Mining, pp. 254–260 (1999)Google Scholar
- 2.Cai, C.H., Fu, A.W.C., Cheng, C.H., Kwong, W.W.: Mining association rules with weighted items. In: IDEAS 1998: Proceedings of the 1998 International Symposium on Database Engineering & Applications, pp. 68–77. IEEE Computer Society, Washington, DC (1998)Google Scholar
- 6.Ramkumar, G.D., Sanjay, R., Tsur, S.: Weighted association rules: Model and algorithm. In: Proc. Fourth ACM Int’l Conf. Knowledge Discovery and Data Mining (1998)Google Scholar
- 7.Roiger, R.J., Geatz, M.W.: Data Mining: A Tutorial Based Primer. Addison Edu. Inc. (2003)Google Scholar